Objective and Background: Implants have become a predictable treatment for the replacement of missing teeth. However, they are not without problems. It is important to understand the relationship between implant failure and patient- and implant-level factors. Therefore, the purpose of this retrospective chart review was to explore patient- and implant-level factors associated with implant failure, defined here as the loss, removal, or scheduled removal of the implant.
Methods: Data were abstracted from charts of patients at the University of Minnesota School of Dentistry. Patient-level variables included current smoking status, gender, age, self-reported diabetes, and history of bisphosphonate use. Implant-level variables included implant brand, location, whether sinus or bone augmentation was done, immediate or delayed implant placement and restoration type. Data were entered into a custom, web-based data file. Implant failure was defined as the loss or planned removal of the fixture. Descriptive statistics were computed for patient- and implant-level variables. Hazard ratios (HR) and p-values were calculated using Cox proportional hazards regression. The relationship between the variables and implant failures were explored in both univariable and multivariable analyses. We analyzed data for all implants and separately for implants placed in the posterior maxilla. Results: Smoking was the only patient-level variable associated with implant failure in the univariable analyses (P=0.0109, HR=2.13). Sinus grafting material significantly affected implant failure in both univariable and multivariable analyses (P=<0.0001). Implant-level variables such as length, jaw region, alveolar ridge grafting, and sinus grafting both prior to and at the time of implant placement were not significantly associated with failures (P=>0.05) in the univariable analysis. In the multivariable analysis, overdentures had a failure rate 2.95 times greater than single crowns. In addition, sinus grafting material (P=<0.0001) and bone grafting material (P=0.006) were also significant in the multivariable analysis for all sites. When limiting the multivariable analysis to posterior maxillary sites only bone grafting material was no longer significant, while both restoration type and sinus grafting material remained statistically significant. Interestingly, timing of ridge augmentation, either prior to or at the time of implant placement did not affect implant failure (HR=0.97).
Conclusions: Overall, implants are a predictable treatment option. Smoking negatively affects implant success. Implants supporting overdentures have higher failure rates when compared to those supporting single or multiple fixed restorations and fixed-detachable appliances. Self-reported diabetes, use of bisphosphonates and timing of ridge augmentation were not associated with implant failure.